Cultures organotypiques placentaires = cultures ex vivo

Une partie des expériences a été menée sur des explants placentaires, modèle qui conserve la structure de la villosité et l’ensemble des types cellulaires placentaires. Ce modèle est considéré de ce fait plus physiologique que celui des cultures cellulaires primaires, et est qualifié d’ex vivo dans la littérature. Plusieurs groupes de villosités choriales sont disséqués à partir d’un même placenta, et mis en culture sur Matrigel®. Les explants bénéficient de 18h d’attachement, puis ils sont traités par de l’hCG (de 0 U.I./ml à 100 U.I./ml) pendant 0 à 48h. A chaque temps de prélèvement, les milieux de culture et les explants sont collectés et conservés à -20°C. EG-VEGF est dosé dans les surnageants par Test ELISA. Pour les explants, une partie est incluse en paraffine pour localiser certaines protéines en Immunohistochimie, alors que l’ARN d’une autre partie des explants est extrait pour étudier l’expression de différents gènes par RT-Q-PCR (voir figure 63).

Dans un premier temps, nous avons vérifié l’expression du récepteur à l’hCG, le LH/CGR, dans la villosité choriale. Puis nous avons étudié l’effet de l’hormone sur l’expression de l’ARNm et la sécrétion d’EG-VEGF. Dans un deuxième temps, nous avons recherché les mécanismes qui régissent l’action de l’hCG sur EG-VEGF. La régulation des PROKRs par l’hCG a également été établie au niveau ARNm et protéine.

Brouillet S. 1,2,3, Hoffmann P. 1,2,4, Chauvet S. 2,3,5 Salomon A.1,2,3, Chamboredon S.1,2, Sergent F.1,2,4, Benharouga M. 2,3,5, Feige JJ.1,2,3, Alfaidy N.1,2,3,4.

1 _{Institut National de la Santé et de la Recherche Médicale, Unité 1036, Laboratoire Biologie du Cancer et de}

l’Infection, Grenoble, France; 2 Commissariat à l’Energie Atomique, iRTSV, Grenoble, France; 3 Université Joseph Fourrier, Grenoble 1, France; 4 Centre Hospitalier Régional Universitaire de Grenoble, Département de Gynécologie, Obstétrique et Médecine de la Reproduction, Grenoble, France; and 5 Centre National de la Recherche Scientifique, UMR 5249, Grenoble, France.

Informative title: New role for hCG in human pregnancy

Correspondance: Dr. Nadia ALFAIDY INSERM U1036

iRTSV/BCI, CEA-G

17, rue des Martyrs. 38054 Grenoble, Cedex 9, France Tel: (33) 438 78 35 01

Fax: (33) 438 78 50 58

e-mail: nadia.alfaidy-benharouga@cea.fr

2

Abstract

EG-VEGF is an angiogenic factor reported to be specific for endocrine tissues, including the placenta. It biological activity is mediated via two G protein-coupled receptors, prokineticin receptor 1 (PROKR1) and prokineticin receptor 2 (PROKR2). We have recently shown that i) EG-VEGF expression peaks between the 10th and 11th week of gestation ii) its mRNA and protein levels are up-regulated by hypoxia iii) EG-VEGF is a negative regulator of trophoblast invasion and iv) its circulating levels are increased in preeclampsia (PE), the most threatening pathology of pregnancy. Here, we investigated the regulation of the expression of EG-VEGF and its receptors by hCG, a key hormone of pregnancy that is also deregulated in PE. During the first trimester of pregnancy, hCG and EG-VEGF exhibit the same pattern of expression suggesting a potential regulation of EG-VEGF by hCG. Both placental explants (PEX) and primary cultures of trophoblasts from the first trimester of pregnancy were used to investigate this hypothesis. Our results show that i) LH/CGR, the hCG receptor, is expressed both in cyto- and syncytiotrophoblasts ii) hCG increases EG-VEGF, PROKR1 and PROKR2 mRNA and protein expression in a dose and time-dependent manners iii) hCG increases the release of EG-VEGF from PEX conditioned media iv) hCG effects are transcriptional and post- transcriptional and v) the hCG effects are mediated by cAMP via cAMP response elements present in the EG-VEGF promoter region. Altogether, these results demonstrate a new role for hCG in the regulation of an emerging regulatory system of placental development, EG-VEGF and its receptors.

The process of embryo implantation and trophoblast invasion is the most limiting factor for successful pregnancy. Molecular interactions at the embryo-maternal interface during the time of adhesion and subsequent invasion are crucial for implantation (1). Failure of these interactions can lead to preeclampsia (PE), early pregnancy loss, or intrauterine growth retardation (IUGR). There is evidence suggesting that cytokines produced by the developing placenta play an important role in these processes (1). We have recently determined the role of a new actor of these processes (2, 3), namely endocrine gland-derived vascular endothelial growth factor (EG- VEGF), also named prokineticin 1(PROK1).

EG-VEGF is a growth factor that was found to be specifically expressed in endocrine tissues including testis, adrenal gland, ovary and placenta (4). It was shown to promote tissue-specific angiogenesis in endocrine organs (5-9). EG-VEGF acts via two G protein-coupled receptors, termed prokineticin receptor 1 (PROKR1) and prokineticin receptor 2 (PROKR2) (10). Recent data from our group reported the expression of EG-VEGF and its receptors in human and mouse placenta and its role during the first trimester of pregnancy (3, 11, 12). We have shown that EG-VEGF is localized to the syncytial layer of the human placenta; that both its expression and that of its receptors are high in this tissue, with the strongest expression peaking between the 10th and 11th weeks of gestation (wg); that EG-VEGF controls trophoblast invasion, and that its circulating levels are significantly elevated in PE (3, 11). More recently, we have also shown that EG-VEGF is a potent angiogenic factor in the placenta, and established its angiogenic role during pregnancy (2). Altogether these findings suggest that EG- VEGF is directly involved in normal placental development and that its expression should be finely regulated. We and others have shown that expression of EG-VEGF and its receptors is up-regulated by hypoxia (4, 11). During placental development, the hypoxic environment lasts from the beginning of implantation to the end of the first trimester. However, the strongest expression of EG-VEGF is between the 8th and 11th wg, suggesting that other factors than hypoxia might regulate the EG-VEGF/ PROKR1/PROKR2 system. To date, little is known about the regulation of EG-VEGF and its receptors, and there is no enlightenment for the peak of expression of EG-VEGF by the end of the first trimester of pregnancy. During the first trimester of pregnancy, one dominant hormone, hCG (human chorionic gonadotropin) exhibits the same pattern of expression to that of EG-VEGF and displays similar effects on placental development (13-15). One of the earliest endocrine roles of hCG is to stimulate the corpus luteum to produce enough progesterone in order to establish pregnancy at the outset. In the placenta, hCG is well known to facilitate trophoblastic differentiation (14, 15), and was reported to

4

induce the expression of specific genes such as vascular endothelial growth factor, leukemia inhibitory factor, and metalloproteinase-9, all central to the establishment of the feto-maternal interface (16, 17). In human placenta, hCG is primarily produced by the syncytiotrophoblast and to a certain extent by the cytotrophoblasts (18, 19). In normal pregnancies, detectable levels of hCG begin to appear in the maternal circulation about 2–3 weeks after conception and the peak is observed at ~8–9 wg before declining significantly in the later stages of pregnancy. High serum hCG levels at mid-late pregnancy have been associated with PE, IUGR and Down's syndrome (DS) (20).

The hCG hormone transduces signals by binding to its specific LH/hCG receptor (LHCGR). Binding of hCG to its receptor generates signal transduction through the activation of the associated heterotrimeric G-proteins and a consequent activation of protein kinase A (PKA) through cAMP mobilization, as well as an increase in intracellular calcium through inositol triphosphate/ phospholipase C pathway (21, 22). LHCGR expression in human placenta has been reported by many groups (23, 24), however, the precise sites of its expression and the type of receptors expressed throughout pregnancy are still under debate (24-28). This discrepancy is probably due to the multiple forms described for this receptor. It has also been reported that in early pregnancy, the LHCGR are truncated and probably non-functional until 9 wg (25, 26, 29).

Because hCG has long been associated with the initiation and maintenance of pregnancy, and since our recent findings propose EG-VEGF as a new factor directly involved in human placentation, we hypothesized that hCG might be involved in the regulation of EG-VEGF and of its receptors. In the present study, we determined, for the first time, the effect of hCG on EG-VEGF secretion; established its effect on EG-VEGF expression both at the mRNA and protein levels; and determined its effect on the expression of PROKR1 and PROKR2. More importantly, we characterized the molecular mechanism by which hCG regulates this new factor and its receptors.

Tissue collection

A total of 48 first-trimester human placentas of 6–11 weeks of gestation (wg) were obtained from elective terminations of pregnancies. Shortly after collection, tissue was fixed in paraformaldehyde at room temperature (for Immunohistochemistry); snap frozen in liquid nitrogen and stored at – 80°C to be used for RNA and protein extraction, or placed in ice-cold Hanks’ balanced salt solution (Ca2-,Mg2- HBSS) and transported to the laboratory for in vitro and ex vivo primary cultures. A total of 29 placentas from 6–11 wg were used for primary cultures. Collection and processing of human placentas was approved by the University Hospital Ethics Committee, and informed consent was obtained from each patient.

Human villous explants cultures

Villous explant cultures were established from first trimester human placentas (6–11 wg). Small fragments of placental villi (15-20 mg wet weight) were placed into 48-well plates precoated with 150 µL per well of diluted Matrigel (matrigel/DMEM-F12) (Becton-Dickinson, le Pont de claix, France) and polymerized at 37°C for 30 min. Explants were cultured in DMEM-Ham’s F-12 (DMEM/F12; Invitrogen, Cergy Pontoise, France) supplemented with 100 µg/mL streptomycin, 100 U/mL penicillin, pH 7.4. After 24h of culture, the medium was changed and explants were incubated in the absence or presence of hCG (10-100 IU/ml) (Sigma Aldrich, France) or forskolin (10µM, Sigma Aldrich) for 0 to 48h. Villous explants were kept in culture for 72h. Explants (from a single placenta) were used in triplicate for each time point. For statistical analysis, the (n) value represents the number of placentas (not explants). In some experiments, trophoblast explants were treated by 50 µg/ml of 5, 6- dichloro-1-β-d-ribofuranosylbenzimidazole (DRB), a potent RNA polymerase inhibitor, or 10 µg/ml cycloheximide (CHX), a translation inhibitor.

Immunohistochemistry

Placental tissues collected at 6–11 wg were fixed for 24 h at 4°C in 4% (vol/vol) paraformaldehyde, embedded in paraffin, and cut into 5 µ m sections as described previously (11). Adjacent sections were stained using specific antibodies and the avidin-biotin immunoperoxidase detection method. Endogenous peroxidase activity was quenched by pretreatment with 3% (vol/vol) hydrogen peroxide in methanol for 30 min. Rabbit polyclonal antibodies developed in our laboratory were used to detect EG-VEGF, PROKR1 and PROKR2 (Covalab, Lyon, France). hCG receptor was detected using the human LHCGR antibody “peptide 28-77” (Santa Cruz Biotechnology). For immunohistochemical detection, antibodies were incubated with the tissue sections for 18 h at 4°C and used at final concentrations of 0.33 µg/ml for anti-EG-VEGF, 0.84 µg/ml for anti-PROKR1, 0.84 µg/ml for anti-PROKR2, 1 µ g/ml for anti-LHCGR. The tissue sections were subsequently washed three times with PBS and incubated with biotinylated goat anti-rabbit IgG (1:250 dilution in blocking solution; Sigma Aldrich, Saint-Quentin Fallavier, France) for 1 h at 4°C. After three PBS washes, the slides were incubated with an avidin biotin complex (Vectastain ABC kit; Vector Laboratories, Burlingame, CA) for 1 h. After a final PBS wash, the immunoreactive proteins were visualized after the addition of 3,3-diaminobenzidine (Dako, Trappes,

6

France) for 2 min and then counterstained with hematoxylin. Control sections were treated with antibodies that had been preabsorbed overnight at 4°C with the appropriate antigen peptides or without primary antibodies.

Western blotting analysis

Frozen placental samples and cultured trophoblast cells were homogenized in RIPA lysis buffer [50 mM Tris- HCl (pH 7.5), 150 mM NaCl, 1% sodium deoxycholate, 0.1% sodium dodecyl sulfate, 1% Triton X-100, 1 mM PMSF, 5 µg/ml leupeptin, 5 µg/ml aprotinin] and processed as previously described (3). Protein extracts were electrophoretically separated on 0.1% sodium dodecyl sulfate-12 % polyacrylamide gels and electrically transferred onto 0.45 µ m nitrocellulose membranes. The membranes were blotted with antibodies against PROKR1, PROKR2 and LHCGR. PROKR1 and PROKR2 antibodies were both used at a final concentration of 0.84 µg/ml, LHCGR antibody was used at 0.75µg/ml. The intensities of immunoreactive bands were measured by scanning the photographic film and analyzing the images on a desktop computer using Scion Image software (version 4.0.2; Scion Corp., Frederick, MD). The mean pixel density for each band was analyzed to obtain relative OD units for PROKR1 and PROKR2 proteins. To standardize for sample loading, the blots were subsequently stripped using a commercially available kit, following the manufacturer's instructions (Re-blot; Millipore) and reprobed with an anti-βactin antibody (Sigma Aldrich, France) as an internal control for protein loading.

EG-VEGF ELISA

EG-VEGF was measured by ELISA (PeproTech, France) in conditioned media from placental explants treated or not with hCG (10-100 IU/ml). Two separated standard curves were constructed to allow accurate readings of samples at upper and lower ranges of the assay. All samples were in the linear range of the standard curves. The detection limit of the assay was 16 pg/mL.

Isolation and treatment of trophoblasts

Placental cytotrophoblasts were isolated from first-trimester human placentas (9–11 wk of gestation, n= 14) and cultured as previously described (30). Briefly, the tissue was thoroughly washed in 50 ml cold sterile HBSS until the supernatant was nearly free of blood. Areas rich in chorionic villi were selected and were minced into small pieces between scalpels. Tissue was incubated in HBSS containing trypsin (0.25%) and DNase (0.2 mg/ml) for 30 min digestions. The dispersed placental cells were filtered through 100 µ m nylon gauze and loaded onto a discontinuous Percoll gradient (5–70% in 5% steps of 3 ml each) and then centrifuged at 1200 x g for 20 min at room temperature to separate the different cell types. Cytotrophoblast cells that sedimented between the density markers of 1.049 and 1.062 g/ml were collected and washed with DMEM. Isolated cells were then incubated with anti-CD9 antibodies and subjected to negative immunomagnetic separation using MiniMacs columns (Miltenyi Biotech, Paris, France). The dispersed trophoblasts were cultured for 24 h at 37°C in 5% CO2 - 95% air to allow attachment. The cells were then divided into two groups: a half was treated one day after the plating to insure the cytotrophoblast phenotype and half was allowed to differentiate into a syncytiotrophoblast phenotype and treated 48h later. Both cultures were treated by hCG (10-100 IU/ml) for 12h (mRNA analysis) or 48h (protein analysis). In some experiments, trophoblast cells were treated by DRB (50 µg/ml), or CHX (10 µg/ml). Cell viability, assessed by Trypan blue exclusion, was more than 95% before and after incubation.

manufacturer's instructions. To remove any genomic DNA contamination, total RNA was treated with RNAse- free DNAse I treatment (Qiagen). Total RNA concentration was determined using Nanodrop. Reverse transcription was performed on 0.5 µg total RNA with Superscript II-RNaseH reverse transcriptase (Invitrogen, Cergy Pontoise, France) under conditions recommended by the manufacturer. EG-VEGF, PROKR1, PROKR2 mRNA and 18S rRNA expression was quantified by real-time RT-PCR using a Light Cycler apparatus (Roche Diagnostics, Meylan, France). The PCR was performed using the primers shown in Table 1 and SYBR green PCR core reagents (LightCycler-FastStart Master SYBR Green I, Roche Diagnostics) according to the manufacturer’s instructions. In addition, several control reactions were routinely run in parallel; this includes RT-PCR run in the absence of reverse transcriptase to confirm the absence of genomic DNA contamination, and reverse transcription reactions without RNA to check for reagent contamination. PCR conditions were: step 1, 94°C for 10 min; step 2, 45 cycles consisting of 95°C for 15 sec, temperature indicated in Table 1 for 5 sec, and 72°C for 10 sec. The results were normalized to 18S rRNA expression levels. To assess linearity and efficiency of PCR amplification, standard curves for all transcripts were generated by using serial dilutions of trophoblast cDNA. A melt curve analysis was carried out on the products of amplification reaction to ascertain the melting temperature of the product.

TABLE 1. Primers used for real-time (RT) PCR

Gene Forward primer (5’–3’) Reverse primer (5’–3’) (°C)

EG-VEGF AGGTCCCCTTCTTCAGGAAACG TCCAGGCTGTGCTCAGGAAAAG 56

PROKR1 GTCCTCGTCATTGTCAAGAGCC AAACACGGTGGGGAAGAAGTCG 58

PROKR2 CATCCCATCGCCTTACTTTGC CTTTTCCTTCACGAACACAGTGG 58

18S TTGTTGGTTTTCGGAACTGAGGC GGCAAATGCTTTCGCTCTGGTC 60

DNA transfection and dual luciferase activity assay

Cos7 cells (100,000 cells/12-well plates) were transfected in Opti-MEM (Invitrogen) using lipofectamine-2000 (Invitrogen) with 0.285 µg pGL3b-luc, 0.500 µg of pGL3-pEG-luc. The PGL3b plasmid was purchased from Promega (Madison, WI, USA) (31).Twenty-four hours after transfection, cells were treated with or without 8- bromo-cAMP (Sigma Aldrich, France) for 4 or 6h. Firefly luciferase activities were measured sequentially with the Dual-Luciferase reporter assay (Promega). Results are expressed as ratios of firefly luciferase activity per µg of proteins.

Mutagenesis and functional analysis

The cAMP binding sites were individually mutated using site-directed mutagenesis. Complementary primers were designed for each mutation. Mutant constructs were generated by PFU turbo (Promega, Madison, WI) PCR using 60 ng of each primer, 20 ng of the pGL3-pEG-luc as template, 250 µM dNTP, 2U PFU turbo polymerase, 1x PFU buffer in a total volume of 50 µl. The recommended cycling conditions were adapted as follows to permit amplification of the full plasmid containing the desired mutation: denaturation 95 °C for 60 s; 18 cycles of 95 °C for 60 s, 55 °C for 60 s, 68 °C for 10 min; final extension 72°C for 10 min. A DpnI digestion was

8

performed prior to bacterial transformation of the plasmid DNA to eliminate the parent promoter plasmid. All oligonucleotides used for these procedures are given in Table 2 (mutations highlighted).

For functional analysis, Cos-7 cells were plated in a 24-well plate the day before transfection in order to achieve a density of 70-80% confluency at the day of the transfection. The cells were transiently transfected with 500 ng pGL3b-pEG-luc expression plasmid or empty pGL3 using lipofectamine-2000 (Invitrogen) according to manufacturer’s instructions. After 48 h, the transfection media was removed and was replaced with fresh media containing or not 500 µM 8-bromo-cAMP and 200 µM IBMX. After 6 hours incubation, the cells were lysed using Passive Lysis Buffer (Promega, Madison, WI) as per the manufacturer’s instructions. The Dual Luciferase Assay System (Promega, Madison, WI) was used to assess the luciferase activity in the cell lysates using a Tecan Infinite M200 microplate reader (Tecan, France). The fold stimulation of the promoter activity was calculated after normalizing the reporter firefly luciferase values to the µg of proteins.

Table 2

Statistical analysis

All data are expressed as mean +/- SE. Statistical comparisons were made using nonparametric test and Mann Whitney test. Calculations were performed using SigmaStat (Jandel Scientific Software, San Rafael,CA).

Expression of LHCGR in human placenta

The first series of experiments were conducted to determine the sites and the levels of expression of hCG receptors in human placenta during the first and third trimester of pregnancy. Fig. 1A, shows representative photographs of LHCGR immunolocalization within the placental villi of human placenta at 8 wg (a) and 10 wg (b). Immunostainings show that LHCGR is highly expressed by cyto - and syncytiotrophoblast cells. A strong expression could also be observed in Hofbauer cells. Negative controls are shown in (c) and (d). Fig. 1B, shows a representative western blot that illustrates the protein levels of LHCGR in human first trimester placenta compared to term placenta. Two major bands with molecular masses of 55 and 70 kDa are observed. The 55 kDa form is known as the precursor form and the 70 kDa as the mature form. A slight decrease in the intensity of the bands is observed towards the end of first trimester. The strongest expression is however observed in term placenta.

hCG effect on EG-VEGF protein expression and secretion

In previous reports from our group (11), we have shown that EG-VEGF is highly expressed by the syncytiotrophoblast layer in the human placenta during the first trimester of pregnancy and that this cytokine is secreted and can be measured in human serum. We therefore investigated the effect of hCG on EG-VEGF expression and secretion in the conditioned media of explant cultures (6 to 11 wg) treated or not with hCG. It was particularly relevant to study hCG effects on EG-VEGF expression in an organotypic system in which villous tissue architecture is maintained. Placental villous explants in culture preserve the topology of intact villi and mimic physiological responses.

hCG effect on EG-VEGF protein expression

Analysis of the hCG effect on EG-VEGF protein expression in placentas from 6 to 11 wg showed differential responses in respect to the gestational age of the placentas examined. The placentas collected from 6-8 wg were less sensitive to hCG treatments compare to those collected from 9-11 wg. Fig 2 shows representative